Jump to Navigation

Intercept or Internet: Assessing Methods for Measuring Changes in Vehicle Miles of Travel

PI: Gian-Claudia Sciara
Funder: ITS MRPI

Growing interest in stemming or slowing climate change by lessening transportation-related greenhouse gas emissions increases the urgency of efforts to reduce the driving, or vehicle-miles of travel (VMT), that people do to meet their daily needs and obligations. In some studies, specific planning strategies such as increasing development density and mixing land uses are associated with lower amounts of VMT. Still, the scientific community has emphatically called for more robust evidence to quantify the relationship between land use and travel behavior (National Research Council, 2009). Public agencies have also noted that limited data hinder efforts to evaluate travel impacts of interventions intended to boost non-vehicular modes. Increases in bicycle and pedestrian volumes and transit ridership would be obvious measures, for example, but “[h]istorically, project sponsors have not been asked to provide before and after data that would even allow for such an evaluation” (Metropolitan Transportation Commission, 2008, p. 12). Indeed, before and after studies measuring whether specific interventions change travel behavior are considered the gold standard, yet they are difficult to do and we have only limited experience with them to guide their design. To inform the methodological deliberations of transportation and behavioral researchers regarding before-and-after data collection options and their pros and cons, this work compares the results of two different surveys for assessing the VMT impacts of a significant land use change in Davis, California. To measure whether the addition of a large, new retail venue impacts the shopping-related VMT of town residents, we conduct both a longer online survey of Davis residents and a lean-and-mean intercept survey administered to shoppers at the retail site. Each method captures a different population, with distinct demographic profiles and travel patterns. We test whether the two data approaches yield consistent estimates of change in town residents’ shopping-VMT, since the addition of the new big box store. We also examine the circumstances that local planners and public agency staff should consider in choosing a method for evaluating VMT change due to a specific new policy or project.

Primary Contact: